OpenAI: gpt-oss-20b

openai/gpt-oss-20b

gpt-oss-20b is a 20B open-weight model, ranking #169 of 235 on the TryAii Score. Its best showing is #22 on MATH. It's served by 16 independent hosts from $0.03/$0.13 per million input/output tokens — about a quarter of what models at its level usually charge.

Prompt $/1M

$0.030

Completion $/1M

$0.130

Context

131,072 tok

Best provider

405 tok/s · Groq

Specs

Model id: openai/gpt-oss-20b
Display name: OpenAI: gpt-oss-20b
Modality: text->text
Tokenizer: GPT
Instruct type: —
Context length: 131,072
Max completion: 117,964
Prompt $/1M: $0.030
Completion $/1M: $0.130
Free tier: no

Capabilities

Function callingno
JSON modeno
System promptno
Visionno
Moderatedno

Benchmark scores

Each axis is one benchmark, normalized to the best model in the field.

No benchmark data for this model.

Speed by provider

Provider	tok/s	TTFT	Total latency
Groqfastest	405	373 ms	510 ms
Amazon Bedrock	362	405 ms	449 ms
WandBfastest	261	241 ms	259 ms
Artificial Analysis	182	566 ms	—
Google	176	355 ms	470 ms
NextBit	149	651 ms	987 ms
Clarifai	137	332 ms	482 ms
Fireworks	105	380 ms	553 ms
Ambient	101	677 ms	1,293 ms
Novita	101	524 ms	571 ms
Together	101	247 ms	312 ms
DeepInfra	85	257 ms	375 ms
CoreWeave	69	152 ms	194 ms
Parasail	51	546 ms	652 ms
Chutes	50	787 ms	787 ms
SiliconFlow	47	1,131 ms	1,239 ms
Phala	47	674 ms	2,045 ms
Darkbloom	22	4,643 ms	5,552 ms
DekaLLM	15	1,443 ms	1,917 ms
AkashML	14	1,783 ms	2,156 ms
openaifastest	—	—	—